{"slug":"operations/30-operational-metrics","title":"30 — Operational Metrics","section":"operations","sectionLabel":"Operations","status":"unclassified","type":"document","path":"operations/30-operational-metrics.mdx","bytes":2264,"description":null,"updated":null,"source":"integratewise/Integratewise-docs-wiki","content":"\n> **Extracted from:** IntegrateWise Platform Specification Suite\n> **Region:** third_region\n> **Original line:** 3421\n> **Lines:** 39 | **Chars:** 1,970\n> **Status:** Raw extraction — requires review and canonicalization\n\n30 — Operational Metrics\n30.1 Why this doc exists\nObservability (16) covers infrastructure health. Operational Metrics covers business health — the leading indicators that are predictive of pipeline, retention, and risk.\n\n30.2 Responsibilities\nDefine, compute, and surface the canonical operational metrics.\nMake every metric queryable per tenant, per workspace, per persona, per agent.\n30.3 Canonical metrics (preserves user’s list)\nWorkspace Activation Rate — % of tenants with at least one wsp* transitioned past bootstrapping.\nTime to First Value (TTFV) — wall clock from OAuth complete → first non-empty projection render.\nConnector Adoption — average number of active conn* per wsp\\_ by industry × department.\nCapability Usage — daily invocations per cap_id, bucketed by persona.\nTwin Acceptance Rate — % of Twin proposals accepted vs surfaced.\nMemory Promotion Rate — L5 → L6 promotions per 24h per tenant.\nSignal Precision — accepted signals / surfaced signals, rolling 7 d.\nApproval Latency — GovernanceRequested → GovernanceApproved p50/p95.\n30.4 Inputs\nAll event bus events (filtered / aggregated).\nTenant onboarding completion events (existing spec).\n30.5 Outputs\nTime-series metrics in a TSDB.\nAggregate reports rolled up nightly.\n30.6 Events produced\nOperationalMetricComputed, OperationalMetricAnomalyDetected.\n30.7 Events consumed\nAll events (filtered by metric rule).\n30.8 APIs\nGET /ops/metrics/{metric_id}, POST /ops/metrics/{metric_id}/compute, GET /ops/dashboards/{tenant_id}.\n30.9 State transitions\npreregistered → live → deprecated. Anomaly state: nominal → warning → critical.\n\n30.10 Anomaly detection\nPer-metric threshold low/high plus weekly ML-driven adjustment; admins can pin.\n30.11 Failure handling\nCompute lag beyond SLO → MetricStale event + dashboard badge.\nAnomaly spam-protection: min interval 15 minutes between alerts per metric per tenant.\n30.12 Extension points\nCustom metrics OP_METRIC(name).\nCustom anomaly rules OP_ANOMALY(name).\n"}